Pub Date : 2025-11-01DOI: 10.1016/j.landig.2025.100924
Arman Koul MS , Deborah Duran PhD , Tina Hernandez-Boussard PhD
In the evolving landscape of artificial intelligence (AI), the assumption that more data lead to better models has driven unchecked reliance on synthetic data to augment training datasets. Although synthetic data address crucial shortages of real-world training data, their overuse might propagate biases, accelerate model degradation, and compromise generalisability across populations. A concerning consequence of the rapid adoption of synthetic data in medical AI is the emergence of synthetic trust—an unwarranted confidence in models trained on artificially generated datasets that fail to preserve clinical validity or demographic realities. In this Viewpoint, we advocate for caution in using synthetic data to train clinical algorithms. We propose actionable safeguards for synthetic medical AI, including standards for training data, fragility testing during development, and deployment disclosures for synthetic origins to ensure end-to-end accountability. These safeguards uphold data integrity and fairness in clinical applications using synthetic data, offering new standards for responsible and equitable use of synthetic data in health care.
{"title":"Synthetic data, synthetic trust: navigating data challenges in the digital revolution","authors":"Arman Koul MS , Deborah Duran PhD , Tina Hernandez-Boussard PhD","doi":"10.1016/j.landig.2025.100924","DOIUrl":"10.1016/j.landig.2025.100924","url":null,"abstract":"<div><div>In the evolving landscape of artificial intelligence (AI), the assumption that more data lead to better models has driven unchecked reliance on synthetic data to augment training datasets. Although synthetic data address crucial shortages of real-world training data, their overuse might propagate biases, accelerate model degradation, and compromise generalisability across populations. A concerning consequence of the rapid adoption of synthetic data in medical AI is the emergence of synthetic trust—an unwarranted confidence in models trained on artificially generated datasets that fail to preserve clinical validity or demographic realities. In this Viewpoint, we advocate for caution in using synthetic data to train clinical algorithms. We propose actionable safeguards for synthetic medical AI, including standards for training data, fragility testing during development, and deployment disclosures for synthetic origins to ensure end-to-end accountability. These safeguards uphold data integrity and fairness in clinical applications using synthetic data, offering new standards for responsible and equitable use of synthetic data in health care.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 11","pages":"Article 100924"},"PeriodicalIF":24.1,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145662423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-01DOI: 10.1016/j.landig.2025.100912
William J Bolton PhD , Richard Wilson MPharm , Prof Mark Gilchrist MSc , Prof Pantelis Georgiou PhD , Prof Alison Holmes MD , Timothy M Rawson PhD
<div><h3>Background</h3><div>Challenges exist when translating artificial intelligence (AI)-driven clinical decision support systems (CDSSs) from research into health-care settings, particularly in infectious diseases, an area in which behaviour, culture, uncertainty, and frequent absence of a ground truth enhance the complexity of medical decision making. We aimed to evaluate clinicians’ perceptions of an AI CDSS for intravenous-to-oral antibiotic switching and how the system influences their decision making.</div></div><div><h3>Methods</h3><div>This randomised, multimethod study enrolled health-care professionals in the UK who were regularly involved in antibiotic prescribing. Participants were recruited through personal networks and the general email list of the British Infection Association. The first part of the study involved a semistructured interview about participants’ experience of antibiotic prescribing and their perception of AI. The second part used a custom web app to run a clinical vignette experiment: each of the 12 case vignettes consisted of a patient currently receiving intravenous antibiotics, and participants were asked to decide whether or not the patient was suitable for switching to oral antibiotics. Participants were assigned to receive either standard of care (SOC) information, or SOC alongside our previously developed AI-driven CDSS and its explanations, for each vignette across two groups. We assessed differences in participant choices according to the intervention they were assigned, both for each vignette and overall; evaluated the aggregate effect of the CDSS across all switching decisions; and characterised the decision diversity across participants. In the third part of the study, participants completed the system usability scale (SUS) and technology acceptance model (TAM) questionnaires to enable their opinions of the AI CDSS to be assessed.</div></div><div><h3>Findings</h3><div>59 clinicians were directly contacted or responded to recruitment emails, 42 of whom from 23 hospitals in the UK completed the study between April 23, 2024, and Aug 16, 2024. The median age of participants was 39 years (IQR 37–47), 19 (45%) were female and 23 (55%) were male, 26 (62%) were consultants and 16 (38%) were training-grade doctors, and 14 (33%) specialised in infectious diseases. Interviews revealed mixed individualisation of prescribing and uneven use of technology, alongside enthusiasm for AI, which was conditional on evidence and usability but constrained by behavioural inertia and infrastructure limitations. Case vignette completion times and many decisions were equivalent between SOC and CDSS interventions, with clinicians able to identify and ignore incorrect advice. When a statistical difference was observed, the CDSS influenced participants towards not switching (χ<sup>2</sup> 7·73, p=0·0054; logistic regression odds ratio 0·13 [95% CI 0·03–0·50]; p=0·0031). AI explanations were used only 9% of the time when available.
{"title":"The impact of artificial intelligence-driven decision support on uncertain antimicrobial prescribing: a randomised, multimethod study","authors":"William J Bolton PhD , Richard Wilson MPharm , Prof Mark Gilchrist MSc , Prof Pantelis Georgiou PhD , Prof Alison Holmes MD , Timothy M Rawson PhD","doi":"10.1016/j.landig.2025.100912","DOIUrl":"10.1016/j.landig.2025.100912","url":null,"abstract":"<div><h3>Background</h3><div>Challenges exist when translating artificial intelligence (AI)-driven clinical decision support systems (CDSSs) from research into health-care settings, particularly in infectious diseases, an area in which behaviour, culture, uncertainty, and frequent absence of a ground truth enhance the complexity of medical decision making. We aimed to evaluate clinicians’ perceptions of an AI CDSS for intravenous-to-oral antibiotic switching and how the system influences their decision making.</div></div><div><h3>Methods</h3><div>This randomised, multimethod study enrolled health-care professionals in the UK who were regularly involved in antibiotic prescribing. Participants were recruited through personal networks and the general email list of the British Infection Association. The first part of the study involved a semistructured interview about participants’ experience of antibiotic prescribing and their perception of AI. The second part used a custom web app to run a clinical vignette experiment: each of the 12 case vignettes consisted of a patient currently receiving intravenous antibiotics, and participants were asked to decide whether or not the patient was suitable for switching to oral antibiotics. Participants were assigned to receive either standard of care (SOC) information, or SOC alongside our previously developed AI-driven CDSS and its explanations, for each vignette across two groups. We assessed differences in participant choices according to the intervention they were assigned, both for each vignette and overall; evaluated the aggregate effect of the CDSS across all switching decisions; and characterised the decision diversity across participants. In the third part of the study, participants completed the system usability scale (SUS) and technology acceptance model (TAM) questionnaires to enable their opinions of the AI CDSS to be assessed.</div></div><div><h3>Findings</h3><div>59 clinicians were directly contacted or responded to recruitment emails, 42 of whom from 23 hospitals in the UK completed the study between April 23, 2024, and Aug 16, 2024. The median age of participants was 39 years (IQR 37–47), 19 (45%) were female and 23 (55%) were male, 26 (62%) were consultants and 16 (38%) were training-grade doctors, and 14 (33%) specialised in infectious diseases. Interviews revealed mixed individualisation of prescribing and uneven use of technology, alongside enthusiasm for AI, which was conditional on evidence and usability but constrained by behavioural inertia and infrastructure limitations. Case vignette completion times and many decisions were equivalent between SOC and CDSS interventions, with clinicians able to identify and ignore incorrect advice. When a statistical difference was observed, the CDSS influenced participants towards not switching (χ<sup>2</sup> 7·73, p=0·0054; logistic regression odds ratio 0·13 [95% CI 0·03–0·50]; p=0·0031). AI explanations were used only 9% of the time when available. ","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 11","pages":"Article 100912"},"PeriodicalIF":24.1,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145726136","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-01DOI: 10.1016/j.landig.2025.100921
Yoni Schirris MSc , Rosie Voorthuis MSc , Mark Opdam BSc , Marte Liefaard MSc , Prof Gabe S Sonke PhD , Gwen Dackus PhD , Vincent de Jong PhD , Yuwei Wang MSc , Annelot Van Rossum PhD , Tessa G Steenbruggen PhD , Lars C Steggink PhD , Elisabeth G E de Vries PhD , Prof Marc van de Vijver PhD , Roberto Salgado PhD , Efstratios Gavves PhD , Prof Paul J van Diest PhD , Prof Sabine C Linn PhD , Jonas Teuwen PhD , Renee Menezes PhD , Marleen Kok PhD , Hugo M Horlings PhD
<div><h3>Background</h3><div>The density of stromal tumour-infiltrating lymphocytes (TILs) is a prognostic factor for patients with triple-negative breast cancer and reflects their immune response. Computational TIL assessment has the potential to assist pathologists in this labour-intensive task, because it can be quick and reproducible. However, computational TIL assessment models heavily rely on detailed annotations and use complex deep learning pipelines that pose challenges for model iterations and clinical deployment. Here, we propose and validate a fundamentally simpler deep learning-based model that is trained in only 10 min on 100 times fewer pathologist annotations.</div></div><div><h3>Methods</h3><div>We collected whole slide images (WSIs) with TIL scores and clinical data of 2340 patients with breast cancer, including 790 patients with triple-negative breast cancer, from three cohorts in three countries (one each in the USA, UK, and Netherlands) and three randomised clinical trials in the Netherlands. Morphological features were extracted from WSIs using a pathology foundation model. Our model, label-efficient computational stromal TIL assessment (ECTIL), directly regresses the WSI TIL score from these features. We trained ECTIL on a single cohort from The Cancer Genome Atlas (n=356, ECTIL-TCGA), on only triple-negative breast cancer samples from four cohorts (n=400, ECTIL-TNBC), and on all molecular subtypes of five cohorts (n=1964, ECTIL-combined). We computed the concordance between ECTIL and the pathologist using the Pearson's correlation coefficient (<em>r</em>) and computed the area under the receiver operating characteristic curve (AUROC) using the pathologist TIL scores split into the clinically relevant TILs-high (≥30%) and TILs-low (<30%) groups. We also performed multivariate Cox regression analyses on the PARADIGM cohort with complete clinicopathological variables (n=384) to assess hazard ratios for overall survival, independent of clinicopathological factors.</div></div><div><h3>Findings</h3><div>ECTIL-TCGA showed concordance with the pathologist over five heterogeneous external cohorts (<em>r</em>=0·54–0·74, AUROC 0·80–0·94). ECTIL-TNBC showed a higher performance than ECTIL-TCGA on the PARADIGM cohort (<em>r</em> 0·64, AUROC 0·83 <em>vs r</em> 0·58, AUROC 0·80), and ECTIL-combined attained the highest concordance on an external test set (<em>r</em> 0·69, AUROC 0·85). Multivariate cox regression analyses indicated that every 10% increase of ECTIL-combined TIL scores was associated with improved overall survival (hazard ratio 0·85, 95% CI 0·77–0·93; p=0·0007), which was independent of clinicopathological variables and similar to the pathologist score (0·86, 0·81–0·92; p<0·0001).</div></div><div><h3>Interpretation</h3><div>In conclusion, our study showed that ECTIL could score TILs on haematoxylin and eosin-stained, formalin-fixed, paraffin-embedded WSIs in a single step, attaining high concordance with an expert pat
{"title":"Label-efficient computational tumour infiltrating lymphocyte assessment in breast cancer (ECTIL): multicentre validation in 2340 patients with breast cancer","authors":"Yoni Schirris MSc , Rosie Voorthuis MSc , Mark Opdam BSc , Marte Liefaard MSc , Prof Gabe S Sonke PhD , Gwen Dackus PhD , Vincent de Jong PhD , Yuwei Wang MSc , Annelot Van Rossum PhD , Tessa G Steenbruggen PhD , Lars C Steggink PhD , Elisabeth G E de Vries PhD , Prof Marc van de Vijver PhD , Roberto Salgado PhD , Efstratios Gavves PhD , Prof Paul J van Diest PhD , Prof Sabine C Linn PhD , Jonas Teuwen PhD , Renee Menezes PhD , Marleen Kok PhD , Hugo M Horlings PhD","doi":"10.1016/j.landig.2025.100921","DOIUrl":"10.1016/j.landig.2025.100921","url":null,"abstract":"<div><h3>Background</h3><div>The density of stromal tumour-infiltrating lymphocytes (TILs) is a prognostic factor for patients with triple-negative breast cancer and reflects their immune response. Computational TIL assessment has the potential to assist pathologists in this labour-intensive task, because it can be quick and reproducible. However, computational TIL assessment models heavily rely on detailed annotations and use complex deep learning pipelines that pose challenges for model iterations and clinical deployment. Here, we propose and validate a fundamentally simpler deep learning-based model that is trained in only 10 min on 100 times fewer pathologist annotations.</div></div><div><h3>Methods</h3><div>We collected whole slide images (WSIs) with TIL scores and clinical data of 2340 patients with breast cancer, including 790 patients with triple-negative breast cancer, from three cohorts in three countries (one each in the USA, UK, and Netherlands) and three randomised clinical trials in the Netherlands. Morphological features were extracted from WSIs using a pathology foundation model. Our model, label-efficient computational stromal TIL assessment (ECTIL), directly regresses the WSI TIL score from these features. We trained ECTIL on a single cohort from The Cancer Genome Atlas (n=356, ECTIL-TCGA), on only triple-negative breast cancer samples from four cohorts (n=400, ECTIL-TNBC), and on all molecular subtypes of five cohorts (n=1964, ECTIL-combined). We computed the concordance between ECTIL and the pathologist using the Pearson's correlation coefficient (<em>r</em>) and computed the area under the receiver operating characteristic curve (AUROC) using the pathologist TIL scores split into the clinically relevant TILs-high (≥30%) and TILs-low (<30%) groups. We also performed multivariate Cox regression analyses on the PARADIGM cohort with complete clinicopathological variables (n=384) to assess hazard ratios for overall survival, independent of clinicopathological factors.</div></div><div><h3>Findings</h3><div>ECTIL-TCGA showed concordance with the pathologist over five heterogeneous external cohorts (<em>r</em>=0·54–0·74, AUROC 0·80–0·94). ECTIL-TNBC showed a higher performance than ECTIL-TCGA on the PARADIGM cohort (<em>r</em> 0·64, AUROC 0·83 <em>vs r</em> 0·58, AUROC 0·80), and ECTIL-combined attained the highest concordance on an external test set (<em>r</em> 0·69, AUROC 0·85). Multivariate cox regression analyses indicated that every 10% increase of ECTIL-combined TIL scores was associated with improved overall survival (hazard ratio 0·85, 95% CI 0·77–0·93; p=0·0007), which was independent of clinicopathological variables and similar to the pathologist score (0·86, 0·81–0·92; p<0·0001).</div></div><div><h3>Interpretation</h3><div>In conclusion, our study showed that ECTIL could score TILs on haematoxylin and eosin-stained, formalin-fixed, paraffin-embedded WSIs in a single step, attaining high concordance with an expert pat","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 11","pages":"Article 100921"},"PeriodicalIF":24.1,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145745067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-01DOI: 10.1016/j.landig.2025.100915
Charles R Cleland , Adnan Tufail , Catherine Egan , Xiaoxuan Liu , Alastair K Denniston , Alicja Rudnicka , Christopher G Owen , Covadonga Bascaran , Matthew J Burton
{"title":"Independent and openly reported head-to-head comparative validation studies of AI medical devices: a necessary step towards safe and responsible clinical AI deployment","authors":"Charles R Cleland , Adnan Tufail , Catherine Egan , Xiaoxuan Liu , Alastair K Denniston , Alicja Rudnicka , Christopher G Owen , Covadonga Bascaran , Matthew J Burton","doi":"10.1016/j.landig.2025.100915","DOIUrl":"10.1016/j.landig.2025.100915","url":null,"abstract":"","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 11","pages":"Article 100915"},"PeriodicalIF":24.1,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145745105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-28DOI: 10.1016/j.landig.2025.100930
{"title":"Correction to Lancet Digital Health 2025; 7: 100866.","authors":"","doi":"10.1016/j.landig.2025.100930","DOIUrl":"https://doi.org/10.1016/j.landig.2025.100930","url":null,"abstract":"","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":" ","pages":"100930"},"PeriodicalIF":24.1,"publicationDate":"2025-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145402261","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.landig.2025.100937
Judy W Gichoya MD , Rogers Mwavu MS , Frank Minja MD , Nadi Kaonga MD PhD , Saptarshi Purkayastha PhD , Janice Newsome MD
In a rural clinic in southwestern Uganda, Dr Sarah examines cervical images on her smartphone, receiving real-time artificial intelligence-powered guidance from a gynaecologic oncologist located hundreds of miles away. Once imaginary, this scenario now represents a highly probable future of digital health innovation transforming cancer care globally. With over 35 million new cases of cancer estimated by 2050, and up to 70% of deaths anticipated to disproportionately occur in low-income and middle-income countries (LMICs), digital solutions can be leveraged to accelerate the closure of these cancer care gaps. The global oncology community has responded to this imminent crisis by proposing several interventions, including promoting workforce education, mentorship, and task shifting; supporting early diagnosis and referrals through integrated diagnostics; prioritising and implementing prevention strategies such as tobacco cessation, cervical cancer screening, and vaccination; standardising and personalising treatment through increased participation in clinical trials and provision of essential cancer medications; and strengthening health-care systems. Across all these strategic pillars, digital health tools are crucial for advancing cancer care and narrowing existing global and geographical disparities in LMICs. In this Series paper, we evaluate the current status of these digital innovations in the context of cancer care.
{"title":"Leveraging digital technologies to reduce cancer disparities in low-income and middle-income countries","authors":"Judy W Gichoya MD , Rogers Mwavu MS , Frank Minja MD , Nadi Kaonga MD PhD , Saptarshi Purkayastha PhD , Janice Newsome MD","doi":"10.1016/j.landig.2025.100937","DOIUrl":"10.1016/j.landig.2025.100937","url":null,"abstract":"<div><div>In a rural clinic in southwestern Uganda, Dr Sarah examines cervical images on her smartphone, receiving real-time artificial intelligence-powered guidance from a gynaecologic oncologist located hundreds of miles away. Once imaginary, this scenario now represents a highly probable future of digital health innovation transforming cancer care globally. With over 35 million new cases of cancer estimated by 2050, and up to 70% of deaths anticipated to disproportionately occur in low-income and middle-income countries (LMICs), digital solutions can be leveraged to accelerate the closure of these cancer care gaps. The global oncology community has responded to this imminent crisis by proposing several interventions, including promoting workforce education, mentorship, and task shifting; supporting early diagnosis and referrals through integrated diagnostics; prioritising and implementing prevention strategies such as tobacco cessation, cervical cancer screening, and vaccination; standardising and personalising treatment through increased participation in clinical trials and provision of essential cancer medications; and strengthening health-care systems. Across all these strategic pillars, digital health tools are crucial for advancing cancer care and narrowing existing global and geographical disparities in LMICs. In this Series paper, we evaluate the current status of these digital innovations in the context of cancer care.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 10","pages":"Article 100937"},"PeriodicalIF":24.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145530955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.landig.2025.100933
Lawrence A Shaktah MS , Zunamys I Carrero PhD , Katherine Jane Hewitt MBChB , Marco Gustav MSc , Matthew Cecchini MD PhD , Sebastian Foersch MD , Sabina Berezowska MD , Prof Jakob Nikolas Kather MD
Artificial intelligence (AI) is on the verge of reshaping cancer diagnostics through integration into digital pathology workflows. Despite the progression of AI towards real-world deployment, challenges in interpretability, validation, and clinical integration persist. AI models support the interpretation of stains including haematoxylin and eosin, enabling tumour classification, grading, and biomarker quantification, with clinical applications for targets such as HER2 and PD-L1. In addition, AI models enable the quantification of subtle microscopic patterns with prognostic and predictive values across tumour types. Herein, we provide an overview of the applications of AI in pathology and address emerging regulatory and ethical considerations. We also discuss the disparities in adoption across care settings and emphasise the importance of validation, human oversight, and post-deployment monitoring for the responsible implementation of AI in pathology-driven workflows. Furthermore, we highlight the technical advancements driving these developments, particularly the transition from hand-crafted machine learning workflows to deep learning, self-supervised learning for foundation models, multimodal models, and agentic AI.
{"title":"Application of artificial intelligence and digital tools in cancer pathology","authors":"Lawrence A Shaktah MS , Zunamys I Carrero PhD , Katherine Jane Hewitt MBChB , Marco Gustav MSc , Matthew Cecchini MD PhD , Sebastian Foersch MD , Sabina Berezowska MD , Prof Jakob Nikolas Kather MD","doi":"10.1016/j.landig.2025.100933","DOIUrl":"10.1016/j.landig.2025.100933","url":null,"abstract":"<div><div>Artificial intelligence (AI) is on the verge of reshaping cancer diagnostics through integration into digital pathology workflows. Despite the progression of AI towards real-world deployment, challenges in interpretability, validation, and clinical integration persist. AI models support the interpretation of stains including haematoxylin and eosin, enabling tumour classification, grading, and biomarker quantification, with clinical applications for targets such as HER2 and PD-L1. In addition, AI models enable the quantification of subtle microscopic patterns with prognostic and predictive values across tumour types. Herein, we provide an overview of the applications of AI in pathology and address emerging regulatory and ethical considerations. We also discuss the disparities in adoption across care settings and emphasise the importance of validation, human oversight, and post-deployment monitoring for the responsible implementation of AI in pathology-driven workflows. Furthermore, we highlight the technical advancements driving these developments, particularly the transition from hand-crafted machine learning workflows to deep learning, self-supervised learning for foundation models, multimodal models, and agentic AI.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 10","pages":"Article 100933"},"PeriodicalIF":24.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145530887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.landig.2025.100948
The Lancet Digital Health
{"title":"Transforming liver care with artificial intelligence","authors":"The Lancet Digital Health","doi":"10.1016/j.landig.2025.100948","DOIUrl":"10.1016/j.landig.2025.100948","url":null,"abstract":"","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 10","pages":"Article 100948"},"PeriodicalIF":24.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145651881","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.landig.2025.100906
Prof Rona Moss-Morris PhD , Prof Christine Norton PhD , Prof Ailsa Hart PhD , Fionn Cléirigh Büttner PhD , Thomas Hamborg MSc , Laura Miller BSc , Imogen Stagg MSc , Prof Qasim Aziz PhD , Wladyslawa Czuber-Dochan PhD , Prof Lesley Dibley PhD , Megan English-Stevens , Julie Flowers , Serena McGuinness MSc , Prof Borislava Mihaylova DPhil , Prof Richard Pollok PhD , Chris Roukas MSc , Prof Sonia Saxena MD , Louise Sweeney PhD , Prof Stephanie Taylor MD , Vari Wileman PhD
Background
Fatigue, pain, and faecal urgency or incontinence are common, debilitating symptoms in inflammatory bowel disease (IBD). We developed IBD-BOOST, a digital, interactive, facilitator-supported, self-management intervention, and aimed to assess its effects compared with care as usual in relieving these symptoms and improving quality of life.
Methods
This multicentre, parallel, randomised controlled trial was conducted online in the UK, with allocation concealment maintained. Participants aged 18 years or older with IBD who rated the impact of fatigue, pain, and faecal urgency or incontinence as 5 or more on a 0–10 scale in a UK national survey were invited. Participants were randomly assigned (1:1) to the online IBD-BOOST programme or care as usual for 6 months via computer-generated randomisation. Primary outcomes were UK Inflammatory Bowel Disease Questionnaire (UK-IBDQ) and Global Rating of Symptom Relief at 6 months post-randomisation. All randomly assigned participants were included in the intention-to-treat and harms analysis. This trial is registered with ISRCTN.com (ISRCTN71618461) and is closed.
Findings
Between Jan 20, 2020, and July 27, 2022, 4449 participants were invited to participate, and 780 participants were randomly assigned: 391 to IBD-BOOST and 389 to care as usual. 524 (67%) of 780 participants were female and 253 (32%) were male. At 6 months, there were no statistically significant differences for UK-IBDQ between the care as usual group (unadjusted mean 62·09 [SD 14·42]) and the IBD-BOOST group (unadjusted mean 60·85 [SD 16·08]; treatment effect estimate: adjusted mean difference –1·67 [95% CI –4·13 to 0·80], p=0·19) or for Global Rating of Symptom Relief (unadjusted mean 3·65 [2·75] vs 4·13 [2·81]; adjusted mean difference 0·44 [95% CI –0·56 to 1·44], p=0·39). Complier-averaged causal effects analysis demonstrated that participants who complied with IBD-BOOST reported lower UK-IBDQ scores than those who would have complied in the care as usual group (mean difference –2·39 [95%CI –4·34 to –0·45], p=0·016). Adverse events and serious adverse events were similar between the IBD-BOOST group (55 [14%] of 391) and care as usual group (79 [20%] of 389). There was one possible treatment-related serious adverse event in the IBD-BOOST group (recurrent sleep disorder) and no deaths.
Interpretation
IBD-BOOST did not statistically significantly improve disease-specific quality of life or Global Rating of Symptom Relief in patients with IBD with fatigue, pain, or faecal urgency or incontinence compared with care as usual. People who complied with the intervention appeared to derive benefit. Future research should focus on enhancing compliance with interventions and targeting them to individuals most likely to benefit.
Funding
UK National Institute for Health and Care Research.
{"title":"Digital cognitive behavioural self-management programme for fatigue, pain, and faecal incontinence in inflammatory bowel disease (IBD-BOOST): a multicentre, parallel, randomised controlled trial","authors":"Prof Rona Moss-Morris PhD , Prof Christine Norton PhD , Prof Ailsa Hart PhD , Fionn Cléirigh Büttner PhD , Thomas Hamborg MSc , Laura Miller BSc , Imogen Stagg MSc , Prof Qasim Aziz PhD , Wladyslawa Czuber-Dochan PhD , Prof Lesley Dibley PhD , Megan English-Stevens , Julie Flowers , Serena McGuinness MSc , Prof Borislava Mihaylova DPhil , Prof Richard Pollok PhD , Chris Roukas MSc , Prof Sonia Saxena MD , Louise Sweeney PhD , Prof Stephanie Taylor MD , Vari Wileman PhD","doi":"10.1016/j.landig.2025.100906","DOIUrl":"10.1016/j.landig.2025.100906","url":null,"abstract":"<div><h3>Background</h3><div>Fatigue, pain, and faecal urgency or incontinence are common, debilitating symptoms in inflammatory bowel disease (IBD). We developed IBD-BOOST, a digital, interactive, facilitator-supported, self-management intervention, and aimed to assess its effects compared with care as usual in relieving these symptoms and improving quality of life.</div></div><div><h3>Methods</h3><div>This multicentre, parallel, randomised controlled trial was conducted online in the UK, with allocation concealment maintained. Participants aged 18 years or older with IBD who rated the impact of fatigue, pain, and faecal urgency or incontinence as 5 or more on a 0–10 scale in a UK national survey were invited. Participants were randomly assigned (1:1) to the online IBD-BOOST programme or care as usual for 6 months via computer-generated randomisation. Primary outcomes were UK Inflammatory Bowel Disease Questionnaire (UK-IBDQ) and Global Rating of Symptom Relief at 6 months post-randomisation. All randomly assigned participants were included in the intention-to-treat and harms analysis. This trial is registered with ISRCTN.com (ISRCTN71618461) and is closed.</div></div><div><h3>Findings</h3><div>Between Jan 20, 2020, and July 27, 2022, 4449 participants were invited to participate, and 780 participants were randomly assigned: 391 to IBD-BOOST and 389 to care as usual. 524 (67%) of 780 participants were female and 253 (32%) were male. At 6 months, there were no statistically significant differences for UK-IBDQ between the care as usual group (unadjusted mean 62·09 [SD 14·42]) and the IBD-BOOST group (unadjusted mean 60·85 [SD 16·08]; treatment effect estimate: adjusted mean difference –1·67 [95% CI –4·13 to 0·80], p=0·19) or for Global Rating of Symptom Relief (unadjusted mean 3·65 [2·75] <em>vs</em> 4·13 [2·81]; adjusted mean difference 0·44 [95% CI –0·56 to 1·44], p=0·39). Complier-averaged causal effects analysis demonstrated that participants who complied with IBD-BOOST reported lower UK-IBDQ scores than those who would have complied in the care as usual group (mean difference –2·39 [95%CI –4·34 to –0·45], p=0·016). Adverse events and serious adverse events were similar between the IBD-BOOST group (55 [14%] of 391) and care as usual group (79 [20%] of 389). There was one possible treatment-related serious adverse event in the IBD-BOOST group (recurrent sleep disorder) and no deaths.</div></div><div><h3>Interpretation</h3><div>IBD-BOOST did not statistically significantly improve disease-specific quality of life or Global Rating of Symptom Relief in patients with IBD with fatigue, pain, or faecal urgency or incontinence compared with care as usual. People who complied with the intervention appeared to derive benefit. Future research should focus on enhancing compliance with interventions and targeting them to individuals most likely to benefit.</div></div><div><h3>Funding</h3><div>UK National Institute for Health and Care Research.</div></div>","PeriodicalId":48534,"journal":{"name":"Lancet Digital Health","volume":"7 10","pages":"Article 100906"},"PeriodicalIF":24.1,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145476647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号